Resume
Shahrin Nakkhatra · Data Scientist, Robi Axiata Ltd.
Professional Summary
Machine Learning Engineer with 5+ years of experience building and deploying end-to-end ML systems, managing data pipelines, and leading AI initiatives. Proven track record in customer analytics, computer vision, and NLP. Seeking tech leadership roles to drive innovation and build high-performing teams.
Experience
Associate Manager, AI & Data Science
Robi Axiata Limited
October 2024 — Present · Dhaka, Bangladesh
- •Developed social media sentiment analysis tool tracking sentiment, intent, emotion, and issues from Facebook comments from Robi, Airtel, Banglalink, and Grameenphone Facebook pages - Built live tableau dashboard presenting insights
- •Created automated SLT KPI monitoring app with database versioning, file-based KPI calculations, UI filtering, and monthly email reporting for senior leadership (Band A-C)
- •Engineered My Robi app inactive user profiling system to segment dormant users and deliver personalized re-engagement offers based on usage patterns
- •Built RAG-based procurement chatbot using financial documents (policies, LOAs, procedures) with vector embeddings for accurate context retrieval and query responses
Machine Learning Data Engineer - Level 2
Optimizely
June 2022 — September 2024 · Remote (US/UK)
- •Built end-to-end customer churn prediction system using time series analysis on product usage data, addressing class imbalance and data quality challenges
- •Architected automated ML pipeline on AWS with monthly retraining, model selection, and deployment using Docker, Kubernetes (Kustomize), and Argo Workflows
- •Implemented CI/CD pipeline via GitHub Actions and managed three-tier data architecture (RAW/STAGING/REPORTING) in Snowflake with DBT transformations
- •Integrated LIME for model explainability and delivered actionable insights to stakeholders through automated reporting dashboard
Machine Learning Engineer
NybSys
October 2021 — May 2022 · Dhaka, Bangladesh
- •Enhanced multi-model surveillance platform (Survello) with anomaly detection, loitering, object counting, mask detection, and motion tracking using multiprocessing
- •Developed real-time drowsiness detection system using facial landmarks and eye aspect ratio analysis with alert mechanisms
- •Led end-to-end development of smartphone-based fall detection app including data collection, synthetic data generation, and on-device ML model deployment
- •Maintained and migrated NybFace IoT-based face recognition API to edge devices for improved latency and offline capability
Global Product Development Officer
Avery Dennison
March 2020 — September 2021 · Dhaka, Bangladesh
- •Managed product lifecycle from quotation to production setup in Oracle database, collaborating with supply chain for material sourcing and product re-engineering
Education
Bachelor of Science in Mechanical Engineering
Bangladesh University of Engineering and Technology (BUET)
July 2014 — April 2019 · Dhaka, Bangladesh
Skills
Languages
PythonMATLABCC++BashSQL
ML/DL Frameworks
PyTorchTensorFlowKeras
MLOps & Cloud
DockerKubernetesGitHub CI/CDAWS (EC2, SageMaker)SnowflakeDBT
Data Science
PandasPySpark
Visualization
MatplotlibPlotlySeabornBokeh
Version Control
Git
API
Flask
Certifications
- •Machine Learning — Stanford University (Coursera): Foundations, algorithms, and mathematics
- •Deep Learning Specialization — Stanford University (Coursera): Neural networks, hyperparameter tuning, optimization